To define certain drug susceptibility rhythms in tumor-free and cancerous mice--studying especially compounds tried in human cancer therapy, with or without timed hormone administration, and using standard endpoints of toxicity (mortality, survival time, hemopoietic depression). Thus, to apply to the treatment of cancers the steadily accumulating methods as well as data of chronobiology, notably from rhythm studies. Thereby: a) to improve the therapeutic ratio of currently available carcinostatic therapy by timing administration according to rhythms known to occur with several frequencies in so- called reference variables reflecting host tolerance and the sensitivity of certain tumors. b) To develop new therapy such as phase-shift and stage-shift drugs (chronobiotics) for the selective manipulation of the rhythms in the tumor and host, with the aim of identifying optimal times for arresting and preferably destroying the cancer--with minimal or no injury to the host. Computer methods developed for interpreting rhythms in extraterrestrial space will be adapted to such clinical purposes. Some computer sub-routines are to be further developed for treating noisy and sparse data such as those that are likely to become available in the clinical situation. Criteria for timing treatment in man in the clinical setting thus are to be derived and tests to be formulated to check out the use of these criteria in timing. With new temporal criteria eventually available, one can aim to synchronize times for treatment that are optimal from several viewpoints. Best timing should satisfy the double aim of treating at a rhythm-determined highest sensitivity of the tumor to the therapy, while this time is synchronized to coincide with the time of maximal host resistance to side effects from the therapy. Stage-shifting or phase-shifting constitutes the third desideratum for chronotherapy.